ENSO Discussion

[High Desert Mat!]

Test

[Phil]

Yeah, like I said there's a huge WWB coming and the EPS forecast has become way more aggressive the past few days. Like I said, good luck predicting an easterly trade burst on the horizon.

 

http://blog.southernwx.com/wp-content/uploads/2018/11/iF38OJ3y-1024x768.jpg

Be careful when extrapolating long-term from an ensemble mean. The state of the intraseasonal forcing at the time of initialization can play heavily depending on the biases of the model in question.

 

In this case, recall the EPS has a fairly prolific WPAC bias that often manifests when intraseasonal forcing is approaching/within the Maritime domain. We saw this occur throughout the summer season.

 

In the near term, we have this to contend with.

 

http://www.atmos.albany.edu/student/ventrice/real_time/timeLon/u.anom.30.5S-5N.gif

[Guest daniel1]

Be careful when extrapolating long-term from an ensemble mean. The state of the intraseasonal forcing at the time of initialization can play heavily depending on the biases of the model in question.

 

In this case, recall the EPS has a fairly prolific WPAC bias that often manifests when intraseasonal forcing is approaching/within the Maritime domain. We saw this occur throughout the summer season.

 

In the near term, we have this to contend with.

 

What would those blues near 180 degrees mean?

[Webberweather53]

Be careful when extrapolating long-term from an ensemble mean. The state of the intraseasonal forcing at the time of initialization can play heavily depending on the biases of the model in question.

 

In this case, recall the EPS has a fairly prolific WPAC bias that often manifests when intraseasonal forcing is approaching/within the Maritime domain. We saw this occur throughout the summer season.

 

In the near term, we have this to contend with.

 

http://www.atmos.albany.edu/student/ventrice/real_time/timeLon/u.anom.30.5S-5N.gif

 

That's the GFS forecast which in the longer term is showing an even stronger WWB than the EPS. I've already discussed the EPS' subsseasonal forcing biases which actually dampened this wave too much in the Pacific and is adjusting to a more amplified event. 

[MossMan]

When do you think this Niño reaches it peak?

It’s going to stay right where it is for the rest of the season...The WHAM told me. 2006/07 repeat here we come!

[Phil]

Also, recall what I said 3 weeks ago about the ongoing episode of destructive interference (which has already brought the SSTAs back down to earth) followed by the next round of constructive (pro niño) interference beginning around Thanksgiving.

 

I also (again) agreed that the sign of the AAM integral in the tropics/subtropics is an ENSO expression.

 

Of course, the integral and spatial balance of the AAM budget reflects a multitude of components (both high and low frequency). The spatial structure of the AAM budget right now (surplus in the low-latitudes, etc) does reflect the slow evolution into +ENSO and the mode of shear stress/momentum deposition associated with the QBO structure as well.

 

However, higher frequency, pseudo-resonant components (IE: cycles of meridional transfer/MT, MJO, etc) also express in the AAM budget/tendencies. And the timing/amplitude of these behaviors also loops in interference with ENSO on both sides of the coin.

 

For the time being, this is an episode of destructive interference. If the current frequency holds, this will dissociate in a few weeks, and we’ll move into a regime of constructive interference towards Thanksgiving.

 

So I think my interpretation was correct here, and I don’t see any reason(s) to change my projection(s).

[Webberweather53]

Also, recall what I said 3 weeks ago about the ongoing episode of destructive interference (which has already brought the SSTAs back down to earth) followed by the next round of constructive (pro niño) interference beginning around Thanksgiving.

 

I also (again) agreed that the sign of the AAM integral in the tropics/subtropics is an ENSO expression.

 

 

 

So I think my interpretation was correct here, and I don’t see any reason(s) to change my projection(s).

 

Yes but you clearly incorrectly assumed the meridional transfer of AAM was not ENSO related when in fact there really hasn't been a meridional transfer to begin with up to this point. There might be after this subseasonal wave

[Phil]

That's the GFS forecast which in the longer term is showing an even stronger WWB than the EPS. I've already discussed the EPS' subsseasonal forcing biases which actually dampened this wave too much in the Pacific and is adjusting to a more amplified event.

Well yes, a period of constructive interference starting around Thanksgiving is what I’ve been predicting for over a month (see my re-quote from Oct 29th on the previous page). But that would still be intraseasonal in nature, not the long-awaited low frequency hydrostatic coupling to the ENSO structure.

[Webberweather53]

That's the GFS forecast which in the longer term is showing an even stronger WWB than the EPS. I've already discussed the EPS' subsseasonal forcing biases which actually dampened this wave too much in the Pacific and is adjusting to a more amplified event. 

 

 

Be careful when extrapolating long-term from an ensemble mean. The state of the intraseasonal forcing at the time of initialization can play heavily depending on the biases of the model in question.

 

In this case, recall the EPS has a fairly prolific WPAC bias that often manifests when intraseasonal forcing is approaching/within the Maritime domain. We saw this occur throughout the summer season.

 

In the near term, we have this to contend with.

 

http://www.atmos.albany.edu/student/ventrice/real_time/timeLon/u.anom.30.5S-5N.gif

 

This WPAC bias that I've shown you time & time again from Kim et al & others clearly denotes that this bias actually enhances the trade winds in the EPS vs reality and that's what's happening here.

[Webberweather53]

Well yes, a period of constructive interference starting around Thanksgiving is what I’ve been predicting for over a month (see my re-quote from Oct 29th on the previous page). But that would still be intraseasonal in nature, not the long-awaited low frequency hydrostatic coupling to the ENSO structure.

 

The intraseasonal forcing support looks stronger than the preceding dampening imo, the net result is enhancement of the ongoing El Nino because again there's asymmetries in the system that enhance WWBs and not easterly trade wind bursts and this is true even if the winds are of the opposite sign and of equivalent magnitude. I'm still waiting on this NINO head fake you've been touting for so long.

[Phil]

This WPAC bias that I've shown you time & time again from Kim et al & others clearly denotes that this bias actually enhances the trade winds in the EPS vs reality and that's what's happening here.

It enhances the trade winds to the east of the dateline. It enhances westerlies to the west of the dateline.

 

Which is (ideally) how you draw from the warm pool and trigger an OKW given the present thermocline structure which remains shoaled west of niño climo.

[Webberweather53]

It enhances the trade winds to the east of the dateline. It enhances westerlies to the west of the dateline.

 

Which is (ideally) how you draw from the warm pool and trigger an OKW given the present thermocline structure which remains shoaled west of niño climo.

 

 

That's not true, you obviously need to read the literature again.

 

Go down to figure 10 and notice the stronger than normal trades thru the dateline and well into the WP vs reality in the EPS (dashed contours).

 

Nice try.

 

https://journals.ametsoc.org/doi/pdf/10.1175/JCLI-D-15-0862.1

[Phil]

The intraseasonal forcing support looks stronger than the preceding dampening imo, the net result is enhancement of the ongoing El Nino because again there's asymmetries in the system that enhance WWBs and not easterly trade wind bursts and this is true even if the winds are of the opposite sign and of equivalent magnitude. I'm still waiting on this NINO head fake you've been touting for so long.

Well, if we fail to achieve official niño status, then it was indeed a “head fake”. I always thought we’d end up with a warm neutral (borderline) result, so whether it ends up warm neutral or weak niño is sort of splitting hairs in the grand scheme of things.

 

The “head fake” comment was more in reference to those predicting a coherent, moderate/strong El Niño event based on climate model projections at the time. What I’ve said is that this will not be a moderate niño, and that inteaseasonal variability will play a larger role than the background state (in terms of the expression of the tropical convective forcing). So far, this has been generally correct, in my opinion.

[Webberweather53]

Well, if we fail to achieve official niño status, then it was indeed a “head fake”. I always thought we’d end up with a warm neutral (borderline) result, so whether it ends up warm neutral or weak niño is sort of splitting hairs in the grand scheme of things.

 

The “head fake” comment was more in reference to those predicting a coherent, moderate/strong El Niño event based on climate model projections at the time. What I’ve said is that this will not be a moderate niño, and that inteaseasonal variability will play a larger role than the background state (in terms of the expression of the tropical convective forcing). So far, this has been generally correct, in my opinion.

 

In Phil's world of make believe where everything is upside down it's a "head fake", but for everyone else with their head screwed on straight and official forecasting agencies it's not but continue living under your rock for all I care instead of admitting you busted, whereas I actually predicted a NINO after the equinox which is indeed occurring.

[Phil]

That's not true, you obviously need to read the literature again.

 

Go down to figure 10 and notice the stronger than normal trades thru the dateline and well into the WP vs reality in the EPS (dashed contours).

 

Nice try.

 

https://journals.ametsoc.org/doi/pdf/10.1175/JCLI-D-15-0862.1

I’ve read that paper, and I assume you’re referring to this image:

 

 

Keep in mind this analysis includes a multitude of ENSO states and is not adjusted for seasonality. The bias in question varies structurally not only due to ENSO but also the seasonal/monsoonal cycle.

[Webberweather53]

I’ve read that paper, and I assume you’re referring to this image:

 

 

Keep in mind this analysis includes a multitude of ENSO states and is not adjusted for seasonality. The bias in question varies structurally based on not only due to ENSO but also the seasonal/monsoonal cycle.

 

Just so we're all on the same page here because it's obvious you're not, those dashed contours are stronger easterly trade winds flowing towards above average precipitation anomalies, and whoa would you look at that there aren't any westerlies west of the dateline like I said. Man what a revelation!

[Webberweather53]

"Still like my idea of a head fake to niño, then a retraction to neutral or warm neutral. The SST warmth should be west-based, with niño-4 possibly running borderline niño conditions through the winter, with niño-3 running neutral, and niño1-2 running cool. The real move to El Niño will start in J/F/M 2019, IMO. "

 

Any other individual aside from Phil reading this would assume you didn't expect an El Nino to form until the winter at the earliest, that doesn't seem to be verifying.

[Phil]

Those dashed contours are easterly trade winds, and whoa would you look at that there aren't any westerlies west of the dateline like I said. Man what a revelation!

Again...the EPS bias varies in longitude and amplitude depending on the ENSO state and the monsoonal cycle/seasonality. The above study does not tune for any of these particular boundary conditions so it has little relevance here (or for any individual year).

 

The hydrostatic/convective biases (which manifest due to errors in the parameterizations of the modeled physics) will naturally bias towards where the convection is actually occurring. So when you have a cold Indo-Pacific and warm WPAC, like in 2018, you can expect the bias to change structurally based on that set of boundary conditions.

[Phil]

"Still like my idea of a head fake to niño, then a retraction to neutral or warm neutral. The SST warmth should be west-based, with niño-4 possibly running borderline niño conditions through the winter, with niño-3 running neutral, and niño1-2 running cool. The real move to El Niño will start in J/F/M 2019, IMO. "

 

Any other individual aside from Phil reading this would assume you didn't expect an El Nino to form until the winter at the earliest, that doesn't seem to be verifying.

Has a Niño formed?

 

Region 3.4 is sitting below niño threshold right now, and the event has been west-based so far. So I’m not sure what’s wrong with that prediction?

[Webberweather53]

Again...the EPS bias varies in longitude and amplitude depending on the ENSO state and seasonality (IE: The Asian monsoonal cycle). The above study does not tune for any of these particular boundary conditions so it has little relevance here.

 

The hydrostatic/convective biases will always bias towards where the convection is actually occurring. So when you have a cold Indo-Pacific and warm WPAC, you can expect the bias to change structurally based on that boundary state.

 

It actually has a lot of relevance, if you think it varies so much to the point where the sign of the anomalies west of the dateline completely reverses and is significant enough to push westerlies nearly up to the dateline then why don't you publish your own study to evaluate their claims because this is merely heresay until that point. This sounds more like an excuse to cover yourself for a misconstrued assumption about the EPS bias enhancing WWBs west of the dateline. Note the bias is particularly prevalent during weak initialized MJO events and lo/behold...

 

Also, if we're going by your logic of relying heavily on when the bias changes, basing your unproven assumption about the westerlies just west of the dateline based on the previous summer's bias in the EPS, would actually suffice to say your assumptions are irrelevant too because we're in boreal winter w/ a significant +ENSO signal.

[Webberweather53]

Has a Niño formed?

 

Region 3.4 is sitting below niño threshold right now, and the event has been west-based so far. So I’m not sure what’s wrong with that prediction?

 

 

NINO region 3.4 SSTs have climbed to nearly +0.5C in ASO, It's sitting above the threshold if you use weekly OISSTv2 data and has been the case for over the past month or so, CDAS tends to be noisy but was of similar amplitude 2 weeks ago @ +1.2C. AAM, SOI, and other metrics also suggest we're entering an El Nino but believe what you want and continue living in your own fantasy world.

[Webberweather53]

Has a Niño formed?

 

Region 3.4 is sitting below niño threshold right now, and the event has been west-based so far. So I’m not sure what’s wrong with that prediction?

 

Oh man it's an astounding 0.03C below the NINO threshold right now definitely clear-cut evidence we're not in an El Nino!!!

[Phil]

It actually has a lot of relevance, if you think it varies so much to the point where the sign of the anomalies west of the dateline completely reverses and is significant enough to push westerlies nearly up to the dateline then why don't you publish your own study to evaluate their claims? This sounds more like an excuse to cover yourself for a misconstrued assumption about the EPS bias enhancing WWBs west of the dateline. Note the bias is particularly prevalent during weak initialized MJO events and lo/behold...

 

Also, if we're going by your logic of relying heavily on when the bias changes, basing your unproven assumption about the westerlies just west of the dateline based on the previous summer's bias in the EPS, would actually suffice to say your assumptions are irrelevant too because we're in boreal winter w/ a significant +ENSO signal.

Unproven assumptions? Oh boy.

 

It’s a question of physics and how they’re parameterized structurally within the model. When you’re talking about constrained hydrostatic/thermodynamic equations, of course the resulting biases/errors will change with the boundary conditions initialized (since the convective structure itself will change, which changes the spatial calculations that produced the bias in the first place). This seems pretty straightforward to me.

 

FWIW, I actually have done my own filtering, for IOD/SIOD and ENSO (along with seasonality) for research purposes. It’s not all that difficult.

[Webberweather53]

I actually have done my own filtering, for IOD/SIOD and ENSO (along with seasonality) for research purposes. It’s not all that difficult.

 

It’s a question of physics and how they’re parameterized structurally within the model. When you’re talking about constrained hydrostatic/thermodynamic equations, of course the resulting biases/errors will change with the boundary conditions initialized (since the convective structure itself will change, which changes the spatial calculations that produced the bias in the first place). This seems pretty straightforward to me.

 

Yes unproven assumptions as in you haven't either shown actual results from the model to support your claims or published it in a peer reviewed article. If it's so straightforward then why don't you prove it's the case because what we're seeing is clearly validating the Kim Maritime Continent bias in the EPS and not what you're trying to blindly assert. Don't be surprised if later forecasts show an even larger WWB. I'll be waiting.

[Webberweather53]

Oh man it's an astounding 0.03C below the NINO threshold right now definitely clear-cut evidence we're not in an El Nino!!!

 

You also conveniently left out the fact that every other region of the equatorial pacific (NINO 1-2, NINO 3, & NINO 4) is above the 0.5C threshold but I'm sure you won't let that get in the way of your poor attempt to cover your own tracks.

[Phil]

If it's so straightforward then why don't you prove it's the case because what we're seeing is clearly validating the Kim Maritime Continent bias

Are you suggesting that model biases remain static even as boundary conditions change? If so, that’s a juggernaut of a claim, my friend.

 

These biases exist in the structure they do based on the relatively stable climatological-thermodynamic boundaries that define the climate system. If/when these conditions change (as they do) the structure of the model bias (assuming it’s a dynamic model) must change as well.

[Webberweather53]

Are you suggesting that model biases remain static even as boundary conditions change? If so, that’s a juggernaut of a claim, my friend.

 

These biases exist in the structure they do based on the climatological boundary conditions in question. If/when they change, they structure of the model bias (assuming it’s a dynamic model) must change as well.

 

I'm suggesting that what's happening now doesn't disprove the findings of the kim study w/ an EPS wet bias in the Maritime Continent.

 

Juggernaut of a claim? You should definitely take a good look in the mirror before touting nonsense like this, you're trying to blindly assert that the changes are so large that the biases in the model completely reverse sign vs climatology at one time of the year and are significant-very significant. Do they change? certainly, but to this extent, yeah we don't know. That's going to take extra analysis to prove. Ball is in your court.

[Phil]

I'm suggesting that what's happening now doesn't disprove the findings of the kim study

 

Juggernaut of a claim? Look in the mirror, you're trying to blindly assert that the changes are so large that the biases in the model completely reverse sign at one time of the year and are significant-very significant. That's going to take extra analysis to prove. Ball is in your court.

Huh? I’m not questioning the results of the study. I’m saying they need to be interpreted correctly and applied in the proper context.

 

The study in question was conducted with the intent of highlighting a dynamic flaw in the model, not to cure one’s own seasonal/subseasonal forecast.

[Phil]

If model biases are stable (as you claim) why didn’t they simply use 1 month or 1 year in their analysis?

 

Seems like that would be much quicker/easier, no?

[Webberweather53]

Huh? I’m not questioning the results of the study. I’m saying they need to be interpreted correctly and applied in the proper context.

 

The study in question was conducted with the intent of highlighting a dynamic flaw in the model, not to cure one’s own seasonal/subseasonal forecast.

 

 

Again, you do realize that the mean bias in the model is in fact stronger easterly trade winds west of the dateline right?

 

What you're clearly suggesting is that these are actually westerly at this time of the year west of the dateline. That's going to take an extra bit of analysis to show on your part because even if the biases in the model change me, you, or anyone else as far as I know is actually fully confident they change THAT much or anywhere to the amount you're trying to suggest.

[Webberweather53]

If model biases are stable (as you claim) why didn’t they simply use 1 month or 1 year in their analysis?

 

Seems like that would be much quicker/easier, no?

 

But even if they did would there be enough internal variability to prove your point? You don't know if there is and have to show that w/ more analysis. Ball is still in your court.

[Phil]

The study specifically targets the MJO cycles, and their corresponding amplitude.

 

Obviously, the MJO structure *at initialization* makes a huge difference in how/to what extent any biases will manifest within the projection period. You have argued this yourself, and correctly so.

 

And knowing that the MJO structure/behavior changes in accordance with ENSO, then clearly, the model biases that manifest via such phenomena will also change in structure.

[Webberweather53]

The study is specifically targeted to MJO cycles, and their corresponding amplitude.

 

Obviously, the MJO structure *at initialization* makes a huge difference in how/to what extent any biases will manifest within the projection period. You have argued this yourself, and correctly so.

 

And if the MJO structure/behavior changes in accordance with ENSO (which it does) then so will the model biases that manifest.

 

Again, none of this actually proves that the model bias at this time of the year and in this ENSO phase are westerly near and west of the dateline like you previously claimed. We know from common sense that they'll be different in some, way, shape or form but we don't know the amplitude you made the leap of faith that it was so large it reversed the sign of the bias and significantly at that.

[Phil]

The findings of this study align perfectly with what I’m saying here.

 

We examined the characteristics of MJO propagation across the Maritime Continent in the ECMWF ensemble prediction system using a 20-yr reforecast dataset produced by the recent operational version (cy40r1). We focused on MJO events initialized over the Indian Ocean (phase 2) and investigated the key dynamical factors attributed to the MJO propagation and prediction

 

Analysis of the initial MJO amplitude and prediction skill relationship shows that the initial amplitude-prediction skill relationship is not linear. Particularly, when the prediction starts with moderate amplitude, predictions can have either high or low skill depending on the initial ocean-atmospheric condition.

Which is why they used 20 years of data (which they also state is possibly not enough, given variations in the background climate state).

[Phil]

And yes, there is a seasonal component. The authors acknowledge this weakness.

 

It’s very clearly stated.

 

While our current study provides some new findings in the area of MJO prediction, there are nonnegligible weaknesses, which are mainly due to the lack of sufficient reforecast sample size and variables. First, the proposed mechanism of the suppressed convection anomaly in the western Pacific playing a role on the eastward propagation is based on existing studies (Hirata et al. 2013; D. Kim et al. 2014) but the limitations in reforecast data availability presented a challenge in exploring detailed mechanisms. Second, we analyzed the MJO prediction irrespective of the season. Generally, the MJO is expected to be better predicted during the boreal winter compared to the summer because of the more pronounced MJO signal (e.g., Wang et al. 2014). The boreal summer intraseasonal oscillation (BSISO) has distinct characteristics that distinguish it from the MJO (e.g., Lawrence and Webster 2002; Kikuchi et al. 2012). Therefore, to accurately evaluate the current state of the MJO (or BSISO) prediction, an appropriate method should be applied to specific season.

[Webberweather53]

And yes, there is a seasonal component. The authors acknowledge this weakness.

 

It’s very clearly stated!

 

 

Nowhere in there did they "very clearly" state that westerlies are enhanced west of the dateline in the EPS bias. 

 

"It enhances the trade winds to the east of the dateline. It enhances westerlies to the west of the dateline."

 

Thanks for playing.

[Webberweather53]

"Particularly, when the prediction starts with moderate amplitude, predictions can have either high or low skill depending on the initial ocean-atmospheric condition."

 

"It enhances westerlies to the west of the dateline."

 

 

To everyone else that has a brain something doesn't add up here, you're full of yourself (as usual).

[Phil]

If you are curious, there are some in-depth analyses that cover longer timespans. Extrapolating via hindcasting could be a useful avenue to decipher the orgins of these biases.

 

Convective heating does indeed appear to be a major source of error, though the parameterization(s) of radiative transfer might be responsible for its inception as a feedback response.

 

https://link.springer.com/article/10.1007/s00382-017-3650-9

 

We focus on the tendency of the simulated intertropical convergence zone in the western equatorial Pacific to drift northwards by between 0.5° and 3° of latitude depending on season. Comparing observations with both fully coupled atmosphere–ocean hindcasts and atmosphere-only hindcasts (driven by observed sea-surface temperatures), we show that the northward drift is caused by a cooling of the sea-surface temperature on the Equator. The cooling is associated with anomalous easterly wind stress and excessive evaporation during the first twenty days of hindcast, both of which occur whether air-sea interactions are permitted or not. The easterly wind bias develops immediately after initialisation throughout the lower troposphere; a westerly bias develops in the upper troposphere after about 10 days of hindcast. At this point, the baroclinic structure of the wind bias suggests coupling with errors in convective heating, although the initial wind bias is barotropic in structure and appears to have an alternative origin.

[Black Hole]

I officially quit trying to explain myself here, we'll just agree to disagree and leave it at that because this is like talking to a brick wall.

Keep posting your content, I always like to read it and learn from it. 

[Phil]

This exchange was extremely disappointing. I feel the cited literature confirms everything I’ve said in this thread to be accurate. I quoted directly from the paper that was supposed to be a refutation.

 

Yet I spent most of the discussion dodging personal attacks and having words put in my mouth. Usually, I find these technical discussions to be interesting and intellectually stimulating, but this was more analogous to a tribal debate on a weenie forum. Highly unpleasant.